Huixuan Tang

Dynamic Graphics Project,
Department of Computer Science,
University of Toronto

email:
hxtang[at]dgp[dot]toronto[dot]edu

office:
BA 5194, 40 St. George Street, Toronto, ON., M5S2E4, Canada


About Me

I'm a PhD. student supervised by Kyros Kutulakos. I also received a MSc. with Kyros in 2010. My research interest is mainly in computational photography and low-level vision.

I have taken research internships at Adobe Research (2014), Qualcomm Canada (through MITACS, 2013-2014), Microsoft Research (2010 in Redmond and 2007-2008 in Beijing) and Motorola China Research Centre (2007-2008).


Recent publications [Google scholar]

Optical aberrations and defocus in photography

High Resolution Photography with an RGB-Infrared Camera
Huixuan Tang, Xiaopeng Zhang, Feng Chen, Shaojie Zhuo, Kiriakos N. Kutulakos and Liang Shen
In Proc. 7th Int. Conf. on Computational Photography (ICCP), Houston, TX, 2015 (Oral)
[paper pdf]

What Does an Aberrated Photo Tell Us about the Lens and the Scene?
Huixuan Tang and Kiriakos N. Kutulakos.
In Proc. 5th Int. Conf. on Computational Photography (ICCP), Boston, MA, 2013. (Oral)
[paper pdf] [slide pdf]
Utilizing Optical Aberrations for Extended-Depth-of-Field Panoramas
Huixuan Tang and Kiriakos N. Kutulakos.
In: Proc. 11th Asian Conf. on Computer Vision (ACCV), Daejeon, Korea, 2012. (Oral)
[paper pdf] [slide pdf]

Blind image quality assessment

Blind Image Quality Assessment using Semi-supervised Rectifier Network
Huixuan Tang, Neel Joshi and Ashish Kapoor.
In Proc. IEEE Conf. on Computer Vision and Pattern Recognition(CVPR), Columbus, OH, 2014. (Poster)
[paper pdf] [poster pdf] [note on code request]


Learning a Blind Measure of Perceptual Image Quality
Huixuan Tang, Neel Joshi and Ashish Kapoor.
In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition(CVPR), Colorado Springs, CO, 2011. (Poster)
[project page] [paper pdf] [poster pdf] [additional results] [US patent] [note on code request]



Course/Previous projects



Rendering realistic lens flare
Evaluating semi-supervised deep belief networks Detail recovery for single image defocus blur
Figure-ground separation